[BEAM-8283] Create a PTransform that reads data from BigQuery in Python SDK

• Type: Improvement
• Status: Resolved
• Priority: P2
• Resolution: Won't Fix
• Fix Version/s: Not applicable
• Component/s: io-py-gcp
• Assignee: Kamil Wasilewski
• Reporter: Kamil Wasilewski
• Created: 19/Sep/19 15:43
• Updated: 16/May/20 14:07
• Resolved: 09/Oct/19 14:22

Description

ATM `BigQuerySource` in Python SDK can only be used by Dataflow runner. Because of this it's impossible to read data from BigQuery on portable runners, such as Flink.

 

The goal is to add a transform, that follows BQ source implementation in Java[1]: use a BigQuery export job to take a snapshot of the table on GCS, and then read from each produced file.

 

[1] https://github.com/apache/beam/blob/master/sdks/java/io/google-cloud-platform/src/main/java/org/apache/beam/sdk/io/gcp/bigquery/BigQuerySourceBase.java